Tracing Thermal Alteration Using Thermal Fluctuation Tomography for Paleointensity Experiments

Abstract:

The Thellier-series experiments [Thellier and Thellier, 1959] are widely considered the most reliable technique to estimate paleointensities, which can provide crucial constraints on the behavior of the geomagnetic field. During the experiments, samples need to be heated multiple times eventually up to the Curie temperature, in order to be thermally demagnetized and to acquire laboratory-applied thermoremanences. However, the thermally induced physical and chemical alteration of magnetic minerals during the stepwise heating treatments can severely bias paleointensity estimates. In this study, we use thermal fluctuation tomography (TFT) [Jackson et al., 2006] to detect the thermal alteration of sister specimens of a lava sample from Galapagos that have been used in a previous paleointensity study [Wang and Kent, 2013], in the hope of reaching a better understanding of how samples alter during Thellier-series paleointensity experiments.

A traditional approach to gauge thermal alteration is to measure hysteresis properties and FORCs before and after heating (Fig. 1a, d). In our study, we used the HT-VSM (a vibrating sample magnetometer equipped with a high-temperature furnace in the Institute for Rock Magnetism) to measure hysteresis loops, back-field direct current demagnetization (DCD) curves (Fig. 1b) at elevated temperatures on specimen GA85.1w. After the first heating cycle (up to 880 K), we repeated the same measurements to gauge thermal alteration (Fig. 1e), which allowed us to observe differences in rock magnetic properties between the first and the second heating cycles. We also plotted the remanent coercivity spectra from the first and the second heating cycles (Fig. 1c, f), and their differences (Fig, 1i) for comparison.

We find that TFT measurements show more details than FORC diagrams of how the sample thermally alters after heating (increase of remanent coercivity in the DCD curves at elevated temperatures), which can be used to determine the qualification or justify the reliability of their paleointensity results (more detailed analysis to be carried out). On the contrary, pTRM checks imbedded in the original Thellier-series experiments (Fig. 1g, h) failed to show signs of alteration [Wang and Kent, 2013], which our comprehensive rock magnetic experiments showed actually happened.